AKS Cluster provisioning state check does not catch the state `succeeded`

[anderl80]

resource "null_resource" "wait_for_aks" {
  depends_on = [azurerm_kubernetes_cluster.k8s]

  provisioner "local-exec" {
    command = <<EOT
      max_retries=10
      retries=0
      while [ "$(az aks show --resource-group ${azurerm_resource_group.rg.name} --name ${azurerm_kubernetes_cluster.k8s.name} --query "provisioningState" -o tsv)" != "Succeeded" ]; do
        if [ $retries -ge $max_retries ]; then
          echo "Max retries exceeded. Exiting..."
          exit 1
        fi
        echo "Waiting for AKS cluster to be fully provisioned... (Attempt: $((retries+1)))"
        retries=$((retries+1))
        sleep 30
      done
    EOT
  }
}

This script runs endlessly for me though the state is "Succeeded".

I modified the script so it echoes the output of the variable:

while [ "$(az aks show --resource-group ${azurerm_resource_group.rg.name} --name ${azurerm_kubernetes_cluster.k8s.name} --query "provisioningState" -o tsv)" != "Succeeded" ]; do
        echo "Current provisioning state: $(az aks show --resource-group ${azurerm_resource_group.rg.name} --name ${azurerm_kubernetes_cluster.k8s.name} --query "provisioningState" -o tsv)"

Which results in:

➜ terraform apply "main.tfplan"  
null_resource.wait_for_aks: Destroying... [id=9197381911910652745]
null_resource.wait_for_aks: Destruction complete after 0s
azurerm_kubernetes_cluster.k8s: Modifying... [id=/subscriptions/b87de7fe-844d-4fd7-aaa7-2754e58c687a/resourceGroups/rg-calm-marlin/providers/Microsoft.ContainerService/managedClusters/cluster-top-gopher]
azurerm_kubernetes_cluster.k8s: Still modifying... [id=/subscriptions/b87de7fe-844d-4fd7-aaa7-...ice/managedClusters/cluster-top-gopher, 10s elapsed]
azurerm_kubernetes_cluster.k8s: Modifications complete after 20s [id=/subscriptions/b87de7fe-844d-4fd7-aaa7-2754e58c687a/resourceGroups/rg-calm-marlin/providers/Microsoft.ContainerService/managedClusters/cluster-top-gopher]
null_resource.wait_for_aks: Creating...
null_resource.wait_for_aks: Provisioning with 'local-exec'...
null_resource.wait_for_aks (local-exec): Executing: ["/bin/sh" "-c" "      max_retries=10\n      retries=0\n      while [ \"$(az aks show --resource-group rg-calm-marlin --name cluster-top-gopher --query \"provisioningState\" -o tsv)\" != \"Succeeded\" ]; do\n        echo \"Current provisioning state: $(az aks show --resource-group rg-calm-marlin --name cluster-top-gopher --query \"provisioningState\" -o tsv)\"\n        if [ $retries -ge $max_retries ]; then\n          echo \"Max retries exceeded. Exiting...\"\n          exit 1\n        fi\n        echo \"Waiting for AKS cluster to be fully provisioned... (Attempt: $((retries+1)))\"\n        retries=$((retries+1))\n        sleep 30\n      done\n"]
null_resource.wait_for_aks (local-exec): WARNING: The behavior of this command has been altered by the following extension: aks-preview
null_resource.wait_for_aks: Still creating... [12s elapsed]
null_resource.wait_for_aks (local-exec): WARNING: The behavior of this command has been altered by the following extension: aks-preview
null_resource.wait_for_aks (local-exec): Current provisioning state: Succeeded
null_resource.wait_for_aks (local-exec): Waiting for AKS cluster to be fully provisioned... (Attempt: 1)
null_resource.wait_for_aks: Still creating... [22s elapsed]